Microsoft Beat Earnings. Then Lost $350 Billion.

Some business news reads like a lullaby.

Morning Brew is the opposite.

A free daily newsletter that breaks down what’s happening in business and culture — clearly, quickly, and with enough personality to keep things interesting.

Each morning brings a sharp, easy-to-read rundown of what matters, why it matters, and what it means to you. Plus, there’s daily brain games everyone’s playing.

Business news, minus the snooze. Read by over 4 million people every morning.

Try Morning Brew for Free

Here's a riddle for you: A company reports $81.3 billion in revenue (beating expectations). Earnings per share of $4.14 (crushing estimates). Cloud business growing 39% year-over-year (still scorching). And yet... the stock drops 12% in six hours, vaporizing $350 billion in market value.

That $350 billion? It's more than the entire market cap of Netflix. Gone before most people finished their morning coffee.

Welcome to the paradox of modern markets: winning isn't enough. You have to win by more than you won last time.

Think of Microsoft like a poker player holding a full house—a great hand by any measure. But here's the catch: they've been betting so aggressively that now they need to draw an even better hand just to break even.

Azure grew 39% this quarter. Sounds incredible, right? But last quarter it grew 42%. That 3-percentage-point slowdown might seem trivial to normal humans, but to Wall Street, it's a five-alarm fire.

Why? Because Microsoft trades at a price-to-earnings ratio of 30. That's not a number that says "steady, reliable growth." That's a number that screams "this company is going to accelerate into the stratosphere." When reality whispers "maybe we're plateauing," the math breaks violently.

CEO Satya Nadella made things worse by mentioning that AI capacity constraints would persist through June 2026. Translation: that 39% growth rate might be a ceiling, not a floor. For a stock priced for acceleration, a ceiling is a death sentence.

Here's where it gets really interesting. While everyone debates Azure growth rates, the real story is hiding in plain sight: Microsoft spent $37.5 billion on capital expenditures this fiscal year. That's a 66% increase.

Let me put that in perspective. Revenue grew 17%. Spending grew 66%. You don't need an MBA to see the problem—they're running faster just to stay in place.

This capex doesn't immediately hurt earnings (it gets spread across the balance sheet), but it absolutely destroys free cash flow. In a single quarter, Microsoft burned roughly 30% of its operating cash flow building AI infrastructure.

And here's the kicker: much of this spending isn't even offensive. It's defensive. Microsoft isn't building data centers because it sees massive opportunity—it's building them because it can't afford to lose cloud market share to Amazon and Google. When everyone in an industry starts spending defensively, nobody wins. It's an arms race where the only guaranteed losers are shareholder returns.

The economics of AI infrastructure might be fundamentally worse than traditional cloud computing. Here's why that matters.

When Amazon built AWS in the early days, they were creating a market. First-mover advantage. Blue ocean. Today, Microsoft is fighting trench warfare for every percentage point of market share against two extremely well-funded rivals.

But it gets worse. Traditional cloud workloads enjoy gross margins of 60-70%. AI is different. Those NVIDIA H100 GPUs powering AI inference cost about $30,000 each, and they become obsolete frighteningly fast as better chips emerge. Early estimates suggest AI margins could be 20-30 percentage points worse than traditional cloud.

So Microsoft is spending more money, competing harder, in a market with worse economics. That's not the setup for a stock trading at 30 times earnings.

When institutional investors see decelerating growth combined with exploding costs, they don't wait around for clarification. They sell first and ask questions later.

Microsoft's stock had run up 60% in less than a year. At those levels, the market was pricing in perfection—flawless execution, accelerating growth, successful AI monetization. When reality delivered "merely excellent" instead of "perfect," the multiple compressed.

A P/E ratio dropping from 30 to 27 doesn't sound dramatic, but it's a 10% haircut on a $3 trillion company. That's how you lose Netflix-sized chunks of value before lunch.

First, stop conflating good companies with good stocks. Microsoft the business is executing beautifully. Microsoft the stock was simply priced for a level of perfection that reality couldn't deliver. These are different things.

Second, earnings beats don't matter if the growth trajectory disappoints. Wall Street doesn't care about what you did—it cares about the derivative, the rate of change. A company can beat every single estimate and still get crushed if it's growing slower than before.

Third, follow the capex. Reported earnings are an accounting construct. Cash flow reveals truth. When a company is burning cash building infrastructure, that tells you something about competitive pressure that income statements hide.

Finally, high valuations create fragility. When you pay for perfection, you get punished for mere excellence. That's not a flaw in the market—it's just math.

Microsoft will probably be fine. It's still one of the best-run companies on earth, with diversified revenue streams, irreplaceable enterprise relationships, and genuine AI capabilities. The business isn't broken.

But if you're concentrated in mega-cap tech—and let's be honest, most of us are through index funds—this violent repricing is worth paying attention to. It's a reminder that valuation always matters eventually, even for the market's darlings.

Sometimes the full house isn't enough. Sometimes you need four of a kind just to stay in the game.

Until next time,

Trade the Times

Disclaimer: This newsletter is for informational purposes only and does not constitute investment advice. Always do your own research before making any financial decisions.